The MPLS WG Archive

[Rahul Aggarwal <rahul@juniper.net>]

On Wed, 10 Sep 2003, Adrian Farrel wrote:

> To summarize the answers I got to this same question when I asked it last month:
> the problem statement will be enhanced to cover packet and non-packet traffic
>

I am in favor of that as we discusssed last month. The problem statement
should be generic.

> Are we due a revision of the problem statement following the discussions from the last
> month?

The intention was to incorporate the problem statement in the requirements
draft, along with the changes.

rahul

>
> I also asked that the ADs and CCAMP be consulted on this work. It is unclear to me that
> - MPLS should work on something that is not packet oriented
> - we should risk a solution in MPLS that does not fit GMPLS
> - we should risk inventing new techniques in MPLS when GMPLS techniques
>   may already exist.
>
> Adrian
>
> ----- Original Message -----
> From: "Loa Andersson" <loa@pi.se>
> To: <dcheng@cisco.com>
> Cc: "'Lou Berger'" <lberger@movaz.com>; <mpls@UU.NET>
> Sent: Tuesday, September 09, 2003 11:04 PM
> Subject: Re: Invitation to discuss: "P2MP problem statement" - decision
>
>
> > Lou and Dean,
> >
> > I guess this is what should be classified as "good question".
> > What Dean says about the possibility to use the P2MP TE LSPs
> > both in packet based and non-packet based networks is true.
> > When it comes to the scope of what the MPLS working group
> > will define it could be another matter, extending this to
> > include specifications that are applicable for non-packet
> > based networks would require that we explicitly says so in
> > the wg charter and that it is OKed by the ADs.
> >
> > Having said this it is my opinion that we should do the work
> > only once and in the mpls wg.
> >
> > /Loa
> >
> > Dean Cheng (dcheng) wrote:
> >
> > > Lou,
> > >
> > >    Yes it does. In fact, the P2MP TE LSPs can be used in
> > >    both packet based (MPLS) or non-packet based (GMPLS)
> > >    networks, as originally intended.
> > >
> > > Dean
> > >
> > > -----Original Message-----
> > > From: owner-mpls@UU.NET [mailto:owner-mpls@UU.NET] On Behalf Of Lou
> > > Berger
> > > Sent: Tuesday, September 09, 2003 12:22 PM
> > > To: Loa Andersson
> > > Cc: mpls@UU.NET
> > > Subject: Re: Invitation to discuss: "P2MP problem statement" - decision
> > >
> > >
> > > Loa,
> > >          Just want to clarify a point:  Do P2MP TE LSPs include all
> > > GMPLS
> > > switching types or just PSC?  I think inclusion of all types is
> > > important/goodness.
> > >
> > > Thanks,
> > > Lou
> > >
> > > At 01:04 PM 9/9/2003, Loa Andersson wrote:
> > >
> > >
> > >>WG,
> > >>
> > >>I have the working group consensus to ask the ADs to
> > >>add the p-2-mp milestones to our charter. In fact we
> > >>have only seen support :).
> > >>
> > >>Alex,
> > >>
> > >>could you start the process to add the following milestone
> > >>the the mpls wg charter.
> > >>
> > >>Milestones
> > >>
> > >>- produce a requirement specification that defines the scope,
> > >>  requirements, and issues to resolve for how to setup a P2MP TE LSPs
> > >>  by Minneapolis 2003 meeting.
> > >>
> > >>- produce a small number of solution draft(s) which specify protocol
> > >>  extensions, enhancements and mechanisms for P2MP TE LSPs by spring
> > >>  2004 IETF meeting.
> > >>
> > >>
> > >>/Loa
> > >>
> > >>--
> > >>Loa Andersson
> > >>
> > >>Mobile          +46 739 81 21 64
> > >>Email           loa@pi.se
> > >>
> > >
> > >
> > >
> > >
> >
> > --
> > Loa Andersson
> >
> > Mobile          +46 739 81 21 64
> > Email           loa@pi.se
> >
> >
> >
>
>